Pixel to Point Cloud Coordinate Transform

Functional Description

This operator converts input pixel coordinates (2D points on an image) and their corresponding depth values (Z-values in the camera coordinate system) to 3D point coordinates (X, Y, Z) in the camera coordinate system, based on the intrinsic matrix and distortion coefficients obtained from camera calibration.

Usage Scenarios

Image Conversion: Convert pixel coordinates obtained from 2D image processing to precise 3D spatial coordinates.

Inputs and Outputs

Input Items

List of pixels: An N x 3 array or list, where each row represents information for one point in the format [Pixel X-coordinate, Pixel Y-coordinate, Z-value of the point in the camera coordinate system].

Camera internal reference: A 3x3 camera intrinsic matrix, typically in the format: , where fx, fy are focal lengths, and cx, cy are principal point coordinates.

Camera distortion: A list containing camera distortion coefficients, typically 5 coefficients (k1, k2, p1, p2, k3).

Output Items

Camera coordinate system point cloud point: The converted N x 3 array or list, where each row represents the 3D coordinates [X, Y, Z] of a point in the camera coordinate system.